Method for directionally optimizing and modulating sour and sweet taste of reconstituted tobacco through high-precision taste analysis and data analysis technology

Abstract

The invention discloses a method for directionally optimizing and modulating the sour and sweet taste of reconstituted tobacco through a high-precision taste analysis and data analysis technology. The method comprises the following steps: screening a microbial strain, pretreating a reconstituted tobacco leaf base, activating microorganisms, preparing a bacterial suspension, co-modulating, mixing and culturing lactobacillus plantarum and saccharomycetes, post-treating the reconstituted tobacco leaves, and performing high-precision taste analysis and data analysis, so as to construct a lactobacillus plantarum-saccharomycetes cross-domain metabolism synergistic system. According to the method, lactobacillus plantarum and saccharomyces cerevisiae are used as functional strains, after the functional strains are inoculated to a reconstituted tobacco base, oscillation fermentation is conducted, and the sour taste, the sweet taste, the sour-sweet balance, the aroma richness and the overall acceptance score are quantified through electronic tongue detection, PCA and PLS correlation analysis in combination with sensory evaluation. According to the process, the sour strength of the reconstituted tobacco is improved by 2-4 units; the highest sweet intensity is improved by 0.03-0.07 unit, and the total sensory acceptance is improved by more than or equal to 0.8 unit; compared with a direct sensory evaluation method, the method has the advantages that the co-modulation effect is more accurate and more comprehensive.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of reconstituted tobacco processing, in particular to a method for high-precision taste analysis and data analysis technology to optimize the sour and sweet taste of reconstituted tobacco. BACKGROUND

[0002] Reconstituted tobacco is widely used in the tobacco industry, and its sensory quality, especially the sour and sweet taste and aroma, has a significant impact on the final product quality. Traditional reconstituted tobacco bases, as the core material for reducing tar and harm in the tobacco industry, generally have defects such as weak flavor, unbalanced sour and sweet taste, and lack of aroma, which cannot meet the needs of consumers for high-quality tobacco products.

[0003] Using microbial methods to regulate the quality of reconstituted tobacco has become one of the effective means. However, single microbial modulation methods have limitations. For example, Lactobacillus plantarum can only enhance the sour taste in one direction, and has weak effect on improving the sweet taste and aroma; although yeast can improve the sweet taste and aroma to some extent, it cannot improve the sour taste. Therefore, there is an urgent need for a method that can accurately analyze and monitor the taste, and cooperatively optimize the modulation of the sour and sweet taste, aroma, and other sensory dimensions of reconstituted tobacco bases, while improving the quality of reconstituted tobacco and expanding its application value. Due to differences in chemical composition, reconstituted tobacco bases from different sources have different effects on regulation and sensory quality improvement. Therefore, a Lactobacillus plantarum-yeast cross-domain metabolic synergy system is constructed, and a high-precision taste analysis and data analysis technology is used to directionally optimize the modulation of the sour and sweet taste of reconstituted tobacco, which has broken through to achieve the bidirectional improvement of the sour and sweet taste of reconstituted tobacco, and has achieved the improvement of the balance of sour and sweet taste and the overall acceptability, which is an urgent goal to be achieved in the industry. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a method for high-precision taste analysis and data analysis technology to directionally optimize the modulation of the sour and sweet taste of reconstituted tobacco, aiming to construct a Lactobacillus plantarum-yeast cross-domain metabolic synergy system, and use the Insent SA402B taste analysis system to accurately analyze the taste, which has broken through to achieve the bidirectional improvement of the sour and sweet taste of reconstituted tobacco, and has achieved the improvement of the balance of sour and sweet taste and the overall acceptability.

[0005] The technical problem to be solved by the present application is solved by the following technical scheme:

[0006] A method for high-precision taste analysis and data analysis technology to directionally optimize the modulation of the sour and sweet taste of reconstituted tobacco, the method comprising the following steps:

[0007] (1) Microbial strain determination: the microbial strains are Lactobacillus plantarum JNC003.001 and yeast JNC001.002 preserved in the China Center for Type Culture Collection;

[0008] (2) Reconstituted tobacco sheet base pretreatment: remove the surface impurities of the reconstituted tobacco sheet base, then cut or crush to a suitable size to facilitate the reaction of microorganisms with the sheet base;

[0009] (3) Microbial activation and preparation of bacterial suspension: activate Lactobacillus plantarum JNC003.001 and yeast JNC001.002 respectively; Lactobacillus plantarum activation uses MRS medium, cultured at 30-35℃ under anaerobic conditions for 18-24h, continuously passaged for 2-3 times to make the strain active, then configure a certain concentration of Lactobacillus plantarum bacterial suspension; yeast activation uses YPD medium, cultured at 28-32℃ under aerobic conditions for 16-22h, passaged for 2-3 times, and configured into a certain concentration of yeast bacterial suspension;

[0010] (4) Mixed co-modulation culture: mix the reconstituted tobacco sheet base treated in step (2) with the Lactobacillus plantarum bacterial suspension and yeast bacterial suspension prepared in step (3) at a mass-volume ratio of 1:5-1:10; place the mixture in a 28-33℃ culture environment, use the aerobic characteristics of yeast in the early stage, and pass in appropriate amount of air for 12-24h to create a micro-anaerobic environment, then continue to culture under facultative anaerobic conditions for 24-36h, and the whole co-culture process is controlled for 36-60h;

[0011] (5) Reconstituted tobacco post-treatment: separate the excess bacterial suspension and metabolic waste liquid from the reconstituted tobacco obtained in step (4); then dry the sheet base to obtain the treated reconstituted tobacco sheet base finished product;

[0012] (6) High-precision taste analysis: mix the reconstituted tobacco sheet base finished product treated in step (5) and deionized water at a mass-volume ratio of 1:30, homogenize, centrifuge, filter with filter paper, and use the supernatant as the sample to be tested to detect the scores of sourness, sweetness, sour-sweet balance, aroma richness, and overall acceptance by Insent SA402B electronic tongue detection combined with sensory evaluation.

[0013] (7) Data analysis: obtain the final analysis data through the data of Insent SA402B electronic tongue detection combined with principal component analysis PCA and PLS correlation analysis.

[0014] Preferably, in the above technical solution, in step (1), the Saccharomyces cerevisiae JNC001.002 is preserved in the China Center for Type Culture Collection, with the preservation number CCTCC NO:M2024636; and the Lactobacillus plantarum JNC003.001 is preserved in the China Center for Type Culture Collection, with the preservation number CCTCC NO:M2024447.

[0015] Preferably, in the above technical solution, in step (2), impurities on the surface of the reconstituted tobacco leaves are removed by air separation, and the leaves are cut into 5cm×5cm squares by shearing blades or crushed into 20-40 mesh particles by mechanical agitator.

[0016] Preferably, in the above technical solution, in step (3), the concentration of the *Lactobacillus plantarum* suspension in step (3) is 10. 6 ~10 8 CFU / mL, yeast suspension concentration 10 6 ~10 8 CFU / mL; Lactobacillus plantarum was activated to bring the strain to an active state, and then a concentration of 10 was prepared. 6 ~10 8 A suspension of *Lactobacillus plantarum* CFU / mL; yeast, after activation, was prepared to a concentration of 10. 6 ~10 8 Yeast suspension at CFU / mL.

[0017] Preferably, in the above technical solution, in step (4), the mass ratio of the reconstituted tobacco leaf base to the volume ratio of the *Lactobacillus plantarum* suspension is 1:5 to 1:10; the mass ratio of the reconstituted tobacco leaf base to the volume ratio of the yeast suspension is 1:5 to 1:10; and the mass ratio of the co-modified reconstituted tobacco leaf base to the volume ratio of the *Lactobacillus plantarum* and yeast suspension is 1:(4 to 6):(4 to 6).

[0018] Preferably, in the above technical solution, in step (4), an appropriate amount of air is introduced, with an air flow rate of 0.5 to 1.5 vvm.

[0019] Preferably, in the above technical solution, in step (5), after the co-preparation and culture are completed, the reconstituted tobacco leaf base containing the culture solution is placed on double-layer Whatman filter paper, and the surface of the leaf base is gently pressed to absorb the excess culture solution using the filter paper; the filter paper is replaced and the operation is repeated 2-4 times until there is no obvious liquid residue on the surface of the leaf base; the filtered leaf base is placed in a centrifuge tube and centrifuged at 3000-5000 r / min for 15-30 min, and the culture solution adsorbed inside the leaf base is detached and precipitated to the bottom of the tube by centrifugal force, thereby achieving solid-liquid separation; after centrifugation, the leaf base is taken out and dried with hot air at a temperature controlled at 40-60℃ until the moisture content is 10%-15%, and the processed reconstituted tobacco leaf base product is obtained.

[0020] Preferably, in the above technical solution, the high-precision taste analysis in step (6) specifically includes:

[0021] The Insent SA402B taste analysis system was used, equipped with a CA0 acid sensor and a CO0 sweet sensor. The test temperature was controlled at 20℃ by circulating water. Before each sample was tested, the following steps were required: positive electrode cleaning for 90 seconds; negative electrode cleaning twice, 120 seconds each time; and reference solution cleaning for 240 seconds.

[0022] The sample testing time for detection and data acquisition is 30 seconds. Data is collected once every 1 second. The data at the 30th second is selected as the valid value. Each sample is tested 4 times. The first cycle is discarded, and the average of the last 3 times is taken as the final taste response value.

[0023] The tastelessness point is defined based on the reference solution: -13 for acidic taste, -6 for salty taste, and 0 for other taste dimensions.

[0024] Preferably, in the above technical solution, the data analysis in step (7) specifically includes:

[0025] The acid and sweet taste response values ​​detected by the electronic tongue were imported into Astree statistical analysis software. Principal components PC1 and PC2 were obtained by PCA dimensionality reduction, and the differences in taste profiles of different samples were presented in the form of scatter plots. Among them, principal component PC1 reflects the difference in acid-sweet balance, and the higher the score, the better the acid-sweet coordination. Principal component PC2 reflects the taste characteristics associated with aroma richness, and the higher the score, the richer the taste substances related to aroma.

[0026] The above-described technical solution of the present invention has the following beneficial effects:

[0027] 1. This invention innovatively constructs a method for targeted optimization and modulation of the sweet and sour taste of reconstituted tobacco leaves using high-precision taste analysis and data parsing technology. This effectively solves the defects of traditional reconstituted tobacco leaves, such as bland flavor, unbalanced sweet and sour taste, and lack of aroma. This method can achieve a breakthrough in improving both the sweet and sour taste of reconstituted tobacco leaves, realizing a balance between sweet and sour taste and improving overall acceptability.

[0028] 2. This invention utilizes a cross-domain metabolic synergistic system constructed from *Lactobacillus plantarum* and yeast, combined with microenvironment regulation and carbon source complementarity utilization technology. A taste analyzer is used to quantify the acidity, sweetness, sweet-sour balance, aroma richness, and overall acceptability scores through sensory evaluation. Compared to traditional single-microorganism modulation techniques, the co-modulation effect of *Lactobacillus plantarum* and yeast is more significant and comprehensive.

[0029] 3. Using the method of this invention, the Insent SA402B taste analysis system can accurately analyze taste, which can increase the acidity and sweetness of four grades of reconstituted tobacco leaves by more than 10 units and improve the overall sensory acceptance by ≥0.8 units. Attached Figure Description

[0030] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments of the invention and, together with their description, serve to explain the principles of the invention.

[0031] Figure 1 Radar chart of the sweet and sour taste and overall acceptability of reconstituted tobacco leaves processed by microbial fermentation. Detailed Implementation

[0032] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention.

[0033] Experimental methods not specified in the examples generally use conventional conditions and conditions described in the manual, or conditions recommended by the manufacturer. The general equipment, materials, reagents, etc., used are all commercially available unless otherwise specified.

[0034] The preservation information of the strains used in this application is as follows:

[0035] JNC001.002 is *Saccharomyces cerevisiae*, accession number CCTCC NO: M2024636, deposited at the China Center for Type Culture Collection (CCTCC), Wuhan University, China, on April 7, 2024. JNC003.001 is *Lactobacillus plantarum*, accession number CCTCC NO: M2024447, deposited at the China Center for Type Culture Collection (CCTCC), Wuhan University, China, on March 11, 2024.

[0036] like Figure 1 The diagram shows a radar chart illustrating the sweet and sour taste and overall acceptability of reconstituted tobacco leaf bases processed by microbial fermentation. Examples 1-4 illustrate four methods for microbial fermentation of reconstituted tobacco leaf bases (from China Tobacco Schweize (Yunnan) Reconstituted Tobacco Co., Ltd.) to process reconstituted tobacco leaves.

[0037] Example 1

[0038] (1) Screening of microbial strains: Samples were collected from the natural fermentation pile of tobacco. After serial dilution, the samples were spread on MRS agar plates (formula: 10.0g peptone, 10.0g beef extract, 5.0g yeast extract, 20.0g glucose, 5.0g sodium acetate, 2.0g triammonium citrate, 0.2g magnesium sulfate, 0.05g manganese sulfate, 1.0g Tween 80, 15.0g agar, and distilled water to a final volume of 1000mL). The plates were incubated in an anaerobic incubator at 30℃ for 48h. Single colonies were picked and identified as *Lactobacillus plantarum* by Gram staining (Gram positive, short rod-shaped) and biochemical tests (able to ferment glucose to produce acid, lactic acid, etc.). The colonies were named JNC003.001 and stored on slant agar for later use.

[0039] Samples were collected from yeast sludge in a brewery, serially diluted, and spread onto YPD agar plates (formula: 10.0g yeast extract, 20.0g peptone, 20.0g glucose, 20.0g agar, diluted to 1000mL with distilled water). The plates were incubated at 28℃ for 48 hours. Single colonies were picked and identified as yeast JNC001.002 by microscopic examination (fungi, oval shape, exhibiting budding) and biochemical tests (able to ferment glucose to produce ethanol, synthesize esters using organic acids, etc.). The samples were then stored on slant agar for later use.

[0040] Both strains are now deposited at the China Center for Type Culture Collection.

[0041] (2) Pretreatment of reconstituted tobacco leaf substrate: Select reconstituted tobacco leaf A1 and classify it by air separation (air flow rate 20000 m³ / h). 3 After removing surface impurities, the film is crushed to about 30 mesh using a mechanical stirrer, and then sieved to remove large particles, resulting in pretreated film base for later use.

[0042] (3) Microbial activation and bacterial suspension preparation: activate the Lactobacillus plantarum and yeast obtained from (1) screening respectively.

[0043] Activation and suspension preparation of *Lactobacillus plantarum*: Single colonies of JNC003.001 were picked from slant culture medium and inoculated into MRS liquid medium. The cultures were incubated in an anaerobic incubator at 35°C for 20 h for the first activation. Then, 1 mL of the activated culture solution was transferred to fresh MRS liquid medium and incubated under the same conditions for 18 h for the second activation, yielding an active *Lactobacillus plantarum* suspension. The suspension was centrifuged at 5000 rpm for 15 min, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline and adjusted to a concentration of 10. 8 Approximately CFU / mL, used as a suspension of Lactobacillus plantarum.

[0044] Yeast activation and suspension preparation: A single colony of JNC001.002 was picked from the slant culture medium and inoculated into YPD liquid medium. The culture was incubated in an aerobic incubator at 30℃ for 18 hours for the first activation. 1 mL of the activated solution was transferred to fresh YPD liquid medium and incubated under the same conditions for 16 hours for the second activation, yielding an active yeast suspension. The suspension was centrifuged at 3000 rpm for 15 minutes, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline and adjusted to a concentration of 10. 8 Approximately CFU / mL, used as a yeast suspension.

[0045] (4) Co-preparation and culture of Lactobacillus plantarum and yeast: Weigh 100g of the reconstituted tobacco leaf base after (2) pretreatment, mix it with 500mL of Lactobacillus plantarum suspension and 500mL of yeast suspension respectively, and place them in a 5L culture container. In the early stage (first 18h), the temperature is controlled at 30℃ and the air flow rate is 1.0vvm to create a facultative anaerobic environment to promote yeast growth and consume oxygen; after 18h, the culture container is sealed and the environment is changed to anaerobic environment, and culture is continued at 30℃ for 30h. The entire co-culture process lasts for 48h.

[0046] (5) Post-processing of reconstituted tobacco: After co-cultivation, the reconstituted tobacco mixture obtained in step (4) is placed on double-layer Whatman filter paper. The filter paper is replaced and the operation is repeated 3 times until there is no obvious liquid residue on the surface of the leaf base. The filtered leaf base is placed in a centrifuge tube and centrifuged at 5000 r / min for 30 min. After centrifugation, the leaf base is removed and dried with hot air at 50℃ until the moisture content is about 10%, to obtain the co-modified reconstituted tobacco leaf base product (denoted as reconstituted tobacco A1).

[0047] The sensory evaluation, based on ISO8586-2012 certification, was quantitatively assessed using a taste analyzer to evaluate acidity, sweetness, sweet-sour balance, aroma richness, and overall acceptability.

[0048] (6) The reconstituted tobacco leaf base product obtained after modulation is homogenized by adding deionized water at a ratio of 5g:150mL, then transferred to a centrifuge bottle, centrifuged at 10000r / min for 15min, and filtered through a double layer of Whatman filter paper. 80mL of the supernatant is used for taste analysis in the Insent SA402B taste analysis system.

[0049] The taste analysis system is equipped with CA0 (acid) and CO0 (sweet) sensors. The test temperature is controlled at 20℃ using circulating water. Before testing each sample, the following steps are required: positive electrode cleaning (mixture of water, KCl, ethanol, and KOH) for 90 seconds; negative electrode cleaning (mixture of water, ethanol, and hydrochloric acid) twice, 120 seconds each time; and reference solution cleaning (mixture of KCl and tartaric acid, simulating artificial saliva) for 240 seconds. For detection and data acquisition, the sample testing time is 30 seconds, with data collected every 1 second. The data from the 30th second is selected as the valid value. Each sample is tested four times. The first cycle is discarded, and the average of the last three cycles is taken as the final taste response value. The tastelessness point is defined according to the reference solution: -13 for acidity (below this value, there is no acidity), -6 for saltiness, and 0 for other taste dimensions (sweet, umami, bitter, etc.) (≤0 indicates no corresponding taste).

[0050] (7) The multidimensional taste response values ​​(sour and sweet) detected by the electronic tongue were imported into Astree statistical analysis software; the complex taste data were transformed into a few principal components (PC1, PC2) through PCA dimensionality reduction, and the taste profile differences of different samples were presented in the form of scatter plots. Principal component 1 (PC1) can reflect the difference in sour and sweet balance, and the higher the score, the better the sour and sweet coordination; principal component 2 (PC2) can reflect the taste characteristics related to aroma richness, and the higher the score, the richer the aroma-related taste substances (such as esters and amino acids).

[0051] To verify the co-modulation effect, the experiment simultaneously set up a Lactobacillus plantarum single-modulation group, a yeast single-modulation group, and a blank group: the single-modulation group only added the corresponding strain suspension, the blank group added an equal volume of sterile physiological saline, and the culture and post-processing conditions of each group were the same as those of the co-modulation group, and were finally used for taste detection.

[0052] Sensory evaluation was conducted using a taste analyzer in accordance with ISO 8586-2012 certification. The calculation results are as follows:

[0053] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco A1 (modified with Lactobacillus plantarum) were -21.36, 0.08, 3.95, 3.98, and 4.01, respectively.

[0054] The scores for sourness, sweetness, sweet-sour balance, aroma richness, and overall acceptability of reconstituted tobacco A1 (prepared with yeast) were -22.04, 0.09, 4.02, 4.10, and 4.09, respectively.

[0055] The scores for sourness, sweetness, sweet-sour balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A1 (co-modified with Lactobacillus plantarum and yeast) were -21.10, 0.10, 4.18, 4.21, and 4.32, respectively.

[0056] The scores for sourness, sweetness, sweet-sour balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A1 (blank) were -25.56, 0.07, 3.52, 3.44, and 3.51, respectively.

[0057] In summary, the acidity value of the co-modified group of reconstituted tobacco A1 (-21.10) was 4.46 units higher than that of the control group (-25.56), indicating a significant increase in acidity intensity; the sweetness value of the co-modified group of reconstituted tobacco A1 (0.10) was 0.03 units higher than that of the control group (0.07).

[0058] Example 2

[0059] (1) Screening of microbial strains: Samples were collected from the natural fermentation pile of tobacco. After serial dilution, the samples were spread on MRS agar plates (formula: 10.0g peptone, 10.0g beef extract, 5.0g yeast extract, 20.0g glucose, 5.0g sodium acetate, 2.0g triammonium citrate, 0.2g magnesium sulfate, 0.05g manganese sulfate, 1.0g Tween 80, 15.0g agar, and distilled water to a final volume of 1000mL). The plates were incubated in an anaerobic incubator at 30℃ for 48h. Single colonies were picked and identified as *Lactobacillus plantarum* by Gram staining (Gram positive, short rod-shaped) and biochemical tests (able to ferment glucose to produce acid, lactic acid, etc.). The colonies were named JNC003.001 and stored on slant agar for later use.

[0060] Samples were collected from yeast sludge in a brewery, serially diluted, and spread onto YPD agar plates (formula: 10.0g yeast extract, 20.0g peptone, 20.0g glucose, 20.0g agar, diluted to 1000mL with distilled water). The plates were incubated at 28℃ for 48 hours. Single colonies were picked and identified as yeast JNC001.002 by microscopic examination (fungi, oval shape, exhibiting budding) and biochemical tests (able to ferment glucose to produce ethanol, synthesize esters using organic acids, etc.). The samples were then stored on slant agar for later use.

[0061] (2) Pretreatment of reconstituted tobacco leaf substrate: Select reconstituted tobacco leaf A2 and classify it by air separation (air flow rate 20000 m³ / h). 3 After removing surface impurities, the film is crushed to about 30 mesh using a mechanical stirrer, and then sieved to remove large particles, resulting in pretreated film base for later use.

[0062] (3) Microbial activation and bacterial suspension preparation: activate the Lactobacillus plantarum and yeast obtained from (1) screening respectively.

[0063] Activation and suspension preparation of *Lactobacillus plantarum*: Single colonies of JNC003.001 were picked from slant culture medium and inoculated into MRS liquid medium. The cultures were incubated in an anaerobic incubator at 35°C for 20 h for the first activation. Then, 1 mL of the activated culture solution was transferred to fresh MRS liquid medium and incubated under the same conditions for 18 h for the second activation, yielding an active *Lactobacillus plantarum* suspension. The suspension was centrifuged at 5000 rpm for 15 min, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline and adjusted to a concentration of 10. 8 Approximately CFU / mL, used as a suspension of Lactobacillus plantarum.

[0064] Yeast activation and suspension preparation: A single colony of JNC001.002 was picked from the slant culture medium and inoculated into YPD liquid medium. The culture was incubated in an aerobic incubator at 30℃ for 18 hours for the first activation. 1 mL of the activated solution was transferred to fresh YPD liquid medium and incubated under the same conditions for 16 hours for the second activation, yielding an active yeast suspension. The suspension was centrifuged at 3000 rpm for 15 minutes, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline. The concentration was adjusted to approximately 10⁸ CFU / mL to obtain the yeast suspension.

[0065] (4) Co-preparation and culture of Lactobacillus plantarum and yeast: Weigh 100g of the reconstituted tobacco leaf base after (2) pretreatment, mix it with 500mL of Lactobacillus plantarum suspension and 500mL of yeast suspension respectively, and place them in a 5L culture container. In the early stage (first 18h), the temperature is controlled at 30℃ and the air flow rate is 1.0vvm to create a facultative anaerobic environment to promote yeast growth and consume oxygen; after 18h, the culture container is sealed and the environment is changed to anaerobic environment, and culture is continued at 30℃ for 30h. The entire co-culture process lasts for 48h.

[0066] (5) Post-processing of reconstituted tobacco: After co-cultivation, the reconstituted tobacco mixture obtained in step (4) is placed on double-layer Whatman filter paper. The filter paper is replaced and the operation is repeated 4 times until there is no obvious liquid residue on the surface of the leaf base. The filtered leaf base is placed in a centrifuge tube and centrifuged at 5000 r / min for 20 min. After centrifugation, the leaf base is removed and dried with hot air at 50℃ until the moisture content is about 10%, to obtain the co-modified reconstituted tobacco leaf base product (denoted as reconstituted tobacco A2).

[0067] The sensory evaluation, based on ISO8586-2012 certification, was quantitatively assessed using a taste analyzer to evaluate acidity, sweetness, sweet-sour balance, aroma richness, and overall acceptability.

[0068] (6) The reconstituted tobacco leaf base product obtained after modulation is homogenized by adding deionized water at a ratio of 5g:150mL, then transferred to a centrifuge bottle, centrifuged at 10000r / min for 15min, and filtered through a double layer of Whatman filter paper. 80mL of the supernatant is used for taste analysis in the Insent SA402B taste analysis system.

[0069] The taste analysis system is equipped with CA0 (acid) and CO0 (sweet) sensors. The test temperature is controlled at 20℃ using circulating water. Before testing each sample, the following steps are required: positive electrode cleaning (mixture of water, KCl, ethanol, and KOH) for 90 seconds; negative electrode cleaning (mixture of water, ethanol, and hydrochloric acid) twice, 120 seconds each time; and reference solution cleaning (mixture of KCl and tartaric acid, simulating artificial saliva) for 240 seconds. For detection and data acquisition, the sample testing time is 30 seconds, with data collected every 1 second. The data from the 30th second is selected as the valid value. Each sample is tested four times. The first cycle is discarded, and the average of the last three cycles is taken as the final taste response value. The tastelessness point is defined according to the reference solution: -13 for acidity (below this value, there is no acidity), -6 for saltiness, and 0 for other taste dimensions (sweet, umami, bitter, etc.) (≤0 indicates no corresponding taste).

[0070] (7) The multidimensional taste response values ​​(sour and sweet) detected by the electronic tongue were imported into Astree statistical analysis software; the complex taste data were transformed into a few principal components (PC1, PC2) through PCA dimensionality reduction, and the taste profile differences of different samples were presented in the form of scatter plots. Principal component 1 (PC1) can reflect the difference in sour and sweet balance, and the higher the score, the better the sour and sweet coordination; principal component 2 (PC2) can reflect the taste characteristics related to aroma richness, and the higher the score, the richer the aroma-related taste substances (such as esters and amino acids).

[0071] To verify the co-modulation effect, the experiment simultaneously set up a Lactobacillus plantarum single-modulation group, a yeast single-modulation group, and a blank group: the single-modulation group only added the corresponding strain suspension, the blank group added an equal volume of sterile physiological saline, and the culture and post-processing conditions of each group were the same as those of the co-modulation group, and were finally used for taste detection.

[0072] Sensory evaluation was conducted using a taste analyzer in accordance with ISO 8586-2012 certification. The calculation results are as follows:

[0073] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco A2 (modified with Lactobacillus plantarum) were -21.38, 0.08, 3.19, 3.95, and 4.01, respectively.

[0074] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco A2 (prepared with yeast) were -21.99, 0.09, 3.89, 3.34, and 4.05, respectively.

[0075] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A2 (co-prepared with Lactobacillus plantarum and yeast) were -21.16, 0.10, 4.02, 4.18, and 4.08, respectively.

[0076] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A2 (blank) were -23.73, 0.03, 3.40, 3.42, and 3.49, respectively.

[0077] In summary, the acidity value of the reconstituted tobacco A2 co-modification group (-21.16) was 2.57 units higher than that of the control group (-23.73), indicating a significant increase in acidity intensity; the sweetness value of the reconstituted tobacco A2 co-modification group (0.10) was 0.07 units higher than that of the control group (0.03).

[0078] Example 3

[0079] (1) Screening of microbial strains: Samples were collected from the natural fermentation pile of tobacco. After serial dilution, the samples were spread on MRS agar plates (formula: 10.0g peptone, 10.0g beef extract, 5.0g yeast extract, 20.0g glucose, 5.0g sodium acetate, 2.0g triammonium citrate, 0.2g magnesium sulfate, 0.05g manganese sulfate, 1.0g Tween 80, 15.0g agar, and distilled water to a final volume of 1000mL). The plates were incubated in an anaerobic incubator at 30℃ for 48h. Single colonies were picked and identified as *Lactobacillus plantarum* by Gram staining (Gram positive, short rod-shaped) and biochemical tests (able to ferment glucose to produce acid, lactic acid, etc.). The colonies were named JNC003.001 and stored on slant agar for later use.

[0080] Samples were collected from yeast sludge in a brewery, serially diluted, and spread onto YPD agar plates (formula: 10.0g yeast extract, 20.0g peptone, 20.0g glucose, 20.0g agar, diluted to 1000mL with distilled water). The plates were incubated at 28℃ for 48 hours. Single colonies were picked and identified as yeast JNC001.002 by microscopic examination (fungi, oval shape, exhibiting budding) and biochemical tests (able to ferment glucose to produce ethanol, synthesize esters using organic acids, etc.). The samples were then stored on slant agar for later use.

[0081] (2) Pretreatment of reconstituted tobacco leaf substrate: Select reconstituted tobacco leaf A3 and classify it by air separation (air flow rate 20000 m³ / h). 3After removing surface impurities, the film is crushed to about 40 mesh using a mechanical stirrer, and then sieved to remove large particles, resulting in pretreated film base for later use.

[0082] (3) Microbial activation and bacterial suspension preparation: activate the Lactobacillus plantarum and yeast obtained from (1) screening respectively.

[0083] Activation and suspension preparation of *Lactobacillus plantarum*: Single colonies of JNC003.001 were picked from slant agar and inoculated into MRS liquid medium. The cultures were incubated in an anaerobic incubator at 30°C for 24 hours for the first activation. Then, 1 mL of the activated culture was transferred to fresh MRS liquid medium and incubated under the same conditions for 18 hours for the second activation, yielding an active *Lactobacillus plantarum* suspension. The suspension was centrifuged at 5000 rpm for 15 minutes, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline and adjusted to a concentration of 10. 8 Approximately CFU / mL, used as a suspension of Lactobacillus plantarum.

[0084] Yeast activation and suspension preparation: A single colony of JNC001.002 was picked from the slant culture medium and inoculated into YPD liquid medium. The culture was incubated in an aerobic incubator at 30℃ for 18 hours for the first activation. 1 mL of the activated solution was transferred to fresh YPD liquid medium and incubated under the same conditions for 16 hours for the second activation, yielding an active yeast suspension. The suspension was centrifuged at 3000 rpm for 15 minutes, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline. The concentration was adjusted to approximately 10⁸ CFU / mL to obtain the yeast suspension.

[0085] (4) Co-preparation and culture of Lactobacillus plantarum and yeast: Weigh 100g of the reconstituted tobacco leaf base after (2) pretreatment, mix it with 500mL of Lactobacillus plantarum suspension and 500mL of yeast suspension respectively, and place them in a 5L culture container. In the early stage (first 18h), the temperature is controlled at 30℃ and the air flow rate is 1.0vvm to create a facultative anaerobic environment to promote yeast growth and consume oxygen; after 18h, the culture container is sealed and the environment is changed to anaerobic environment, and culture is continued at 30℃ for 30h. The entire co-culture process lasts for 48h.

[0086] (5) Post-processing of reconstituted tobacco: After co-cultivation, the reconstituted tobacco mixture obtained in step (4) is placed on double-layer Whatman filter paper. The filter paper is replaced and the operation is repeated 3 times until there is no obvious liquid residue on the surface of the leaf base. The filtered leaf base is placed in a centrifuge tube and centrifuged at 5000 r / min for 15 min. After centrifugation, the leaf base is taken out and dried with hot air at 50℃ until the moisture content is about 10%, to obtain the co-modified reconstituted tobacco leaf base product (denoted as reconstituted tobacco A3).

[0087] The sensory evaluation, based on ISO8586-2012 certification, was quantitatively assessed using a taste analyzer to evaluate acidity, sweetness, sweet-sour balance, aroma richness, and overall acceptability.

[0088] (6) The reconstituted tobacco leaf base product obtained after modulation is homogenized by adding deionized water at a ratio of 5g:150mL, then transferred to a centrifuge bottle, centrifuged at 10000r / min for 15min, and filtered through a double layer of Whatman filter paper. 80mL of the supernatant is used for taste analysis in the Insent SA402B taste analysis system.

[0089] The taste analysis system is equipped with CA0 (acid) and CO0 (sweet) sensors. The test temperature is controlled at 20℃ using circulating water. Before testing each sample, the following steps are required: positive electrode cleaning (mixture of water, KCl, ethanol, and KOH) for 90 seconds; negative electrode cleaning (mixture of water, ethanol, and hydrochloric acid) twice, 120 seconds each time; and reference solution cleaning (mixture of KCl and tartaric acid, simulating artificial saliva) for 240 seconds. For detection and data acquisition, the sample testing time is 30 seconds, with data collected every 1 second. The data from the 30th second is selected as the valid value. Each sample is tested four times. The first cycle is discarded, and the average of the last three cycles is taken as the final taste response value. The tastelessness point is defined according to the reference solution: -13 for acidity (below this value, there is no acidity), -6 for saltiness, and 0 for other taste dimensions (sweet, umami, bitter, etc.) (≤0 indicates no corresponding taste).

[0090] (7) The multidimensional taste response values ​​(sour and sweet) detected by the electronic tongue were imported into Astree statistical analysis software; the complex taste data were transformed into a few principal components (PC1, PC2) through PCA dimensionality reduction, and the taste profile differences of different samples were presented in the form of scatter plots. Principal component 1 (PC1) can reflect the difference in sour and sweet balance, and the higher the score, the better the sour and sweet coordination; principal component 2 (PC2) can reflect the taste characteristics related to aroma richness, and the higher the score, the richer the aroma-related taste substances (such as esters and amino acids).

[0091] To verify the co-modulation effect, the experiment simultaneously set up a Lactobacillus plantarum single-modulation group, a yeast single-modulation group, and a blank group: the single-modulation group only added the corresponding strain suspension, the blank group added an equal volume of sterile physiological saline, and the culture and post-processing conditions of each group were the same as those of the co-modulation group, and were finally used for taste detection.

[0092] Sensory evaluation was conducted using a taste analyzer in accordance with ISO 8586-2012 certification. The calculation results are as follows:

[0093] The scores for acidity, sweetness, sweet-acid balance, aroma richness, and overall acceptability of reconstituted tobacco A3 (modified with Lactobacillus plantarum) were -21.53, 0.07, 3.76, 3.87, and 3.89, respectively.

[0094] The scores for acidity, sweetness, sweet-sour balance, aroma richness, and overall acceptability of reconstituted tobacco A3 (prepared with yeast) were -22.16, 0.08, 3.82, 3.98, and 3.96, respectively.

[0095] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A3 (co-modified with Lactobacillus plantarum and yeast) were -21.19, 0.09, 3.97, 4.09, and 4.15, respectively.

[0096] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A3 (blank) were -23.51, 0.04, 3.30, 3.41, and 3.46, respectively.

[0097] In summary, the acidity value of the reconstituted tobacco A3 co-modification group (-21.19) was 2.32 units higher than that of the control group (-23.51); the sweetness value of the reconstituted tobacco A3 co-modification group (0.09) was 0.05 units higher than that of the control group (0.04).

[0098] Example 4

[0099] (1) Screening of microbial strains: Samples were collected from the natural fermentation pile of tobacco. After serial dilution, the samples were spread on MRS agar plates (formula: 10.0g peptone, 10.0g beef extract, 5.0g yeast extract, 20.0g glucose, 5.0g sodium acetate, 2.0g triammonium citrate, 0.2g magnesium sulfate, 0.05g manganese sulfate, 1.0g Tween 80, 15.0g agar, and distilled water to a final volume of 1000mL). The plates were incubated in an anaerobic incubator at 30℃ for 48h. Single colonies were picked and identified as *Lactobacillus plantarum* by Gram staining (Gram positive, short rod-shaped) and biochemical tests (able to ferment glucose to produce acid, lactic acid, etc.). The colonies were named JNC003.001 and stored on slant agar for later use.

[0100] Samples were collected from yeast sludge in a brewery, serially diluted, and spread onto YPD agar plates (formula: 10.0g yeast extract, 20.0g peptone, 20.0g glucose, 20.0g agar, diluted to 1000mL with distilled water). The plates were incubated at 28℃ for 48 hours. Single colonies were picked and identified as yeast JNC001.002 by microscopic examination (fungi, oval shape, exhibiting budding) and biochemical tests (able to ferment glucose to produce ethanol, synthesize esters using organic acids, etc.). The samples were then stored on slant agar for later use.

[0101] (2) Pretreatment of reconstituted tobacco leaf base: Select reconstituted tobacco leaf A4 and classify it by air separation (air flow rate 20000 m³ / h). 3After removing surface impurities, the film is crushed to about 40 mesh using a mechanical stirrer, and then sieved to remove large particles, resulting in pretreated film base for later use.

[0102] (3) Microbial activation and bacterial suspension preparation: activate the Lactobacillus plantarum and yeast obtained from (1) screening respectively.

[0103] Activation and suspension preparation of *Lactobacillus plantarum*: Single colonies of JNC003.001 were picked from slant agar and inoculated into MRS liquid medium. The cultures were incubated in an anaerobic incubator at 30°C for 24 hours for the first activation. Then, 1 mL of the activated culture was transferred to fresh MRS liquid medium and incubated under the same conditions for 18 hours for the second activation, yielding an active *Lactobacillus plantarum* suspension. The suspension was centrifuged at 5000 rpm for 15 minutes, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline and adjusted to a concentration of 10. 8 Approximately CFU / mL, used as a suspension of Lactobacillus plantarum.

[0104] Yeast activation and suspension preparation: A single colony of JNC001.002 was picked from the slant culture medium and inoculated into YPD liquid medium. The culture was incubated in an aerobic incubator at 30℃ for 18 hours for the first activation. 1 mL of the activated solution was transferred to fresh YPD liquid medium and incubated under the same conditions for 16 hours for the second activation, yielding an active yeast suspension. The suspension was centrifuged at 3000 rpm for 15 minutes, the supernatant was discarded, and the suspension was resuspended in sterile physiological saline. The concentration was adjusted to approximately 10⁸ CFU / mL to obtain the yeast suspension.

[0105] (4) Co-preparation and culture of Lactobacillus plantarum and yeast: Weigh 100g of the reconstituted tobacco leaf base after (2) pretreatment, mix it with 500mL of Lactobacillus plantarum suspension and 500mL of yeast suspension respectively, and place them in a 5L culture container. In the early stage (first 18h), the temperature is controlled at 30℃ and the air flow rate is 1.0vvm to create a facultative anaerobic environment to promote yeast growth and consume oxygen; after 18h, the culture container is sealed and the environment is changed to anaerobic environment, and culture is continued at 30℃ for 30h. The entire co-culture process lasts for 48h.

[0106] (5) Post-processing of reconstituted tobacco: After co-cultivation, the reconstituted tobacco mixture obtained in step (4) is placed on double-layer Whatman filter paper. The filter paper is replaced and the operation is repeated 3 times until there is no obvious liquid residue on the surface of the leaf base. The filtered leaf base is placed in a centrifuge tube and centrifuged at 5000 r / min for 20 min. After centrifugation, the leaf base is removed and dried with hot air at 45℃ until the moisture content is about 10%, to obtain the co-modified reconstituted tobacco leaf base product (denoted as reconstituted tobacco A4).

[0107] The sensory evaluation, based on ISO8586-2012 certification, was quantitatively assessed using a taste analyzer to evaluate acidity, sweetness, sweet-sour balance, aroma richness, and overall acceptability.

[0108] (6) The reconstituted tobacco leaf base product obtained after modulation is homogenized by adding deionized water at a ratio of 5g:150mL, then transferred to a centrifuge bottle, centrifuged at 10000r / min for 15min, and filtered through a double layer of Whatman filter paper. 80mL of the supernatant is used for taste analysis in the Insent SA402B taste analysis system.

[0109] The taste analysis system is equipped with CA0 (acid) and CO0 (sweet) sensors. The test temperature is controlled at 20℃ using circulating water. Before testing each sample, the following steps are required: positive electrode cleaning (mixture of water, KCl, ethanol, and KOH) for 90 seconds; negative electrode cleaning (mixture of water, ethanol, and hydrochloric acid) twice, 120 seconds each time; and reference solution cleaning (mixture of KCl and tartaric acid, simulating artificial saliva) for 240 seconds. For detection and data acquisition, the sample testing time is 30 seconds, with data collected every 1 second. The data from the 30th second is selected as the valid value. Each sample is tested four times. The first cycle is discarded, and the average of the last three cycles is taken as the final taste response value. The tastelessness point is defined according to the reference solution: -13 for acidity (below this value, there is no acidity), -6 for saltiness, and 0 for other taste dimensions (sweet, umami, bitter, etc.) (≤0 indicates no corresponding taste).

[0110] (7) The multidimensional taste response values ​​(sour and sweet) detected by the electronic tongue were imported into Astree statistical analysis software; the complex taste data were transformed into a few principal components (PC1, PC2) through PCA dimensionality reduction, and the taste profile differences of different samples were presented in the form of scatter plots. Principal component 1 (PC1) can reflect the difference in sour and sweet balance, and the higher the score, the better the sour and sweet coordination; principal component 2 (PC2) can reflect the taste characteristics related to aroma richness, and the higher the score, the richer the aroma-related taste substances (such as esters and amino acids).

[0111] To verify the co-modulation effect, the experiment simultaneously set up a Lactobacillus plantarum single-modulation group, a yeast single-modulation group, and a blank group: the single-modulation group only added the corresponding strain suspension, the blank group added an equal volume of sterile physiological saline, and the culture and post-processing conditions of each group were the same as those of the co-modulation group, and were finally used for taste detection.

[0112] Sensory evaluation was conducted using a taste analyzer in accordance with ISO 8586-2012 certification. The calculation results are as follows:

[0113] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco A4 (modified with Lactobacillus plantarum) were -23.06, 0.08, 3.65, 3.75, and 3.85, respectively.

[0114] The scores for acidity, sweetness, sweet-sour balance, aroma richness, and overall acceptability of reconstituted tobacco A4 (prepared with yeast) were -23.80, 0.09, 3.71, 3.86, and 3.91, respectively.

[0115] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A4 (co-prepared with Lactobacillus plantarum and yeast) were -22.78, 0.11, 3.85, 3.97, and 4.13, respectively.

[0116] The scores for acidity, sweetness, acid-sweet balance, aroma richness, and overall acceptability of reconstituted tobacco leaf A4 (blank) were -25.32, 0.05, 3.2, 3.38, and 3.45, respectively.

[0117] In summary, the acidity value of the reconstituted tobacco A4 co-modification group (-22.78) was 2.54 units higher than that of the control group (-25.32); the sweetness value of the reconstituted tobacco A4 co-modification group (0.11) was 0.06 units higher than that of the control group (0.05).

[0118] Verification Example 1

[0119] This test was conducted by a sensory evaluation panel of 10 sensory experts using methods certified by ISO 8586-2012. Evaluations were based on acidity intensity, sweetness intensity, sweet-acid balance, aroma richness, and overall acceptability (on a 0-10 scale). Average scores were calculated, and a sensory evaluation radar chart was generated (e.g., ...). Figure 1 As shown in the figure, the sensory differences before and after co-modulation and after different film base treatments are compared.

[0120] Four tobacco leaf substrates with different chemical compositions (reconstituted tobacco leaf A1, A2, A3, and A4) showed consistent sensory improvement after co-modification, indicating that the co-metabolistic network of *Lactobacillus plantarum* and yeast has broad substrate adaptability and can degrade various tobacco polymers (cellulose, pectin, protein, etc.). The co-modification process provides stable "flavor defect compensation" for different raw materials. This makes this method suitable for promotion as a standardized pretreatment technology, eliminating the need for complex adjustments based on raw material differences and reducing production process complexity and cost.

[0121] Although the present invention has been disclosed above with reference to embodiments, it is not intended to limit the present invention. Any person skilled in the art can make various different choices and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention is defined by the claims and their equivalents.

Claims

1. A method for targeted optimization and modulation to recreate the sweet and sour taste of tobacco leaves using high-precision taste analysis and data parsing technology, characterized in that, The method includes the following steps: (1) Identification of microbial strains: The microbial strains are Lactobacillus plantarum JNC003.001 and yeast JNC001.002, which are preserved at the China Center for Type Culture Collection; (2) Pretreatment of reconstituted tobacco leaf base: Remove impurities from the surface of the reconstituted tobacco leaf base, and then cut or crush it to a suitable size to facilitate full contact and reaction between microorganisms and the leaf base; (3) Microbial activation and bacterial suspension preparation: Lactobacillus plantarum JNC003.001 and yeast JNC001.002 were activated respectively. Lactobacillus plantarum was activated using MRS medium and cultured at 30-35℃ under anaerobic conditions for 18-24 hours. It was passaged 2-3 times to keep the strain active and then prepared into a bacterial suspension of a certain concentration. Yeast was activated using YPD medium and cultured at 28-32℃ under aerobic conditions for 16-22 hours. It was passaged 2-3 times and then prepared into a bacterial suspension of a certain concentration. (4) Co-culture: The reconstituted tobacco leaf base treated in step (2) is mixed with the plant lactobacillus suspension and yeast suspension prepared in step (3) at a mass-volume ratio of 1:5 to 1:

10. The mixture is placed in a culture environment of 28 to 33°C. In the early stage, the aerobic characteristics of yeast are utilized and an appropriate amount of air is introduced for 12 to 24 hours to create a micro-anaerobic environment. After that, the culture is switched to facultative anaerobic conditions for 24 to 36 hours. The entire co-culture process is controlled to last 36 to 60 hours. (5) Post-processing of reconstituted tobacco leaves: The reconstituted tobacco leaves obtained in step (4) are post-processed to separate the excess bacterial suspension and metabolic waste liquid; then the leaf base is dried to obtain the processed reconstituted tobacco leaf base product. (6) High-precision taste analysis: The reconstituted tobacco leaf base product after step (5) and deionized water are mixed at a mass-volume ratio of 1:30 and homogenized. After centrifugation, the mixture is filtered through filter paper. The supernatant is used as the test sample and is tested by Insent SA402B electronic tongue detection combined with sensory evaluation to quantify sourness, sweetness, sweet-sour balance, aroma richness, and overall acceptability score. (7) Data analysis: The final analysis data is obtained by combining the data detected by the Insent SA402B electronic tongue with principal component analysis (PCA) and PLS correlation analysis.

2. The method according to claim 1, characterized in that, In step (1), Saccharomyces cerevisiae JNC001.002 was deposited at the China Center for Type Culture Collection (CCTCC) with accession number CCTCC NO: M2024636; Lactobacillus plantarum JNC003.001 was deposited at the China Center for Type Culture Collection (CCTCC) with accession number CCTCC NO: M2024447.

3. The method according to claim 1, characterized in that, In step (2), impurities on the surface of the reconstituted tobacco leaves are removed by air separation, and the leaves are cut into 5cm×5cm squares by shearing blades or crushed into 20-40 mesh particles by mechanical agitator.

4. The method according to claim 1, characterized in that, In step (3), the concentration of the *Lactobacillus plantarum* suspension in step (3) is 10. 6 ~10 8 CFU / mL, yeast suspension concentration 10 6 ~10 8 CFU / mL; Lactobacillus plantarum was activated to bring the strain to an active state, and then a concentration of 10 was prepared. 6 ~10 8 A suspension of *Lactobacillus plantarum* CFU / mL; yeast, after activation, was prepared to a concentration of 10. 6 ~10 8 Yeast suspension at CFU / mL.

5. The method according to claim 1, characterized in that, In step (4), the mass ratio of the reconstituted tobacco leaf base to the volume ratio of the *Lactobacillus plantarum* suspension is 1:5 to 1:10; the mass ratio of the reconstituted tobacco leaf base to the volume ratio of the yeast suspension is 1:5 to 1:10; and the mass ratio of the co-modified reconstituted tobacco leaf base to the volume ratio of the *Lactobacillus plantarum* and yeast suspensions is 1:(4 to 6):(4 to 6).

6. The method according to claim 1, characterized in that, In step (4), an appropriate amount of air is introduced, with an air flow rate of 0.5 to 1.5 vvm.

7. The method according to claim 1, characterized in that, In step (5), after the co-cultivation is completed, the reconstituted tobacco leaf base containing the culture solution is placed on a double layer of Whatman filter paper, and the surface of the leaf base is gently pressed to absorb the excess culture solution using the filter paper. The filter paper is replaced and the operation is repeated 2-4 times until there is no obvious liquid residue on the surface of the leaf base. The filtered leaf base is placed in a centrifuge tube and centrifuged at 3000-5000 r / min for 15-30 min. The culture solution adsorbed inside the leaf base is detached and precipitated to the bottom of the tube by centrifugal force, thereby achieving solid-liquid separation. After centrifugation, the leaf base is taken out and dried with hot air at a temperature controlled at 40-60℃ until the moisture content is 10%-15%, thus obtaining the processed reconstituted tobacco leaf base product.

8. The method according to claim 1, characterized in that, The high-precision taste analysis in step (6) specifically includes: The Insent SA402B taste analysis system was used, equipped with a CA0 acid sensor and a CO0 sweet sensor. The test temperature was controlled at 20℃ by circulating water. Before each sample was tested, the following steps were required: positive electrode cleaning for 90 seconds; negative electrode cleaning twice, 120 seconds each time; and reference solution cleaning for 240 seconds. The sample testing time for detection and data acquisition is 30 seconds. Data is collected once every 1 second. The data at the 30th second is selected as the valid value. Each sample is tested 4 times. The first cycle is discarded, and the average of the last 3 times is taken as the final taste response value. The tastelessness point is defined based on the reference solution: -13 for acidic taste, -6 for salty taste, and 0 for other taste dimensions.

9. The method according to claim 1, characterized in that, The data analysis in step (7) specifically involves: The acid and sweet taste response values ​​detected by the electronic tongue were imported into Astree statistical analysis software. Principal components PC1 and PC2 were obtained by PCA dimensionality reduction, and the differences in taste profiles of different samples were presented in the form of scatter plots. Among them, principal component PC1 reflects the difference in acid-sweet balance, and the higher the score, the better the acid-sweet coordination. Principal component PC2 reflects the taste characteristics associated with aroma richness, and the higher the score, the richer the taste substances related to aroma.

Images